Cancer cells reprogram their metabolism to fuel anabolic processes required for their proliferation and survival. One way by which cancer cells reprogram metabolism is by hijacking the evolutionarily conserved metabolic function of the PI3K/Akt/mTORC1 signaling pathway. Another way is by markedly elevating the expression of the hexokinase isoform, HK2, which catalyzes the first committed step in glucose metabolism. The long-term goal of this grant application is to overcome challenges in targeting cancer metabolism and Akt for cancer therapy. Although cancer cells can be selectively detected because of their high glucose metabolism (FDG-PET scan), exploiting this property for selective targeting is challenging because interference with glucose metabolism could have adverse consequences. We overcame this roadblock by showing that hexokinase 2, which catalyzes the first committed step in glycolysis, and is selectively expressed in cancer cells, can be systemically deleted in mice without any adverse consequences. The proposal is based on findings made in my laboratory at the cellular and organismal levels, and address the following paradigm shifts: (i) Germ-line deletion of hexokinase 2, which is markedly elevated in cancer cells, is embryonic lethal. However, we found that its systemic deletion in adult mice is well tolerated, and therapeutic in several mouse models of cancer. This grant application will specifically address the role of hexokinase 2 in metastasis. We will investigate the role of hexokinase 2 in EMT and metastasis through a novel mechanism. (ii) Akt is perhaps the most frequently activated oncoprotein in human cancer. However, we found that, paradoxically, hepatic deletion of Akt1 and Akt2 in mice induces liver injury, inflammation, and early onset of hepatocellular carcinoma. We will determine how the HCC cells survive and proliferate in the absence of Akt. (iii) We found that systemic deletion of Akt1 and Akt2 in adult mice elicits rapid mortality, and that Akt2 deficiency in mice could be pro-tumorigenic and increased metastasis of chemically- induced HCC, possibly because of hyperinsulinemia. Therefore, we launched a comprehensive approach in which Akt1 or Akt2 or both can be conditionally deleted either in a cell autonomous manner or systemically, after tumor onset, to follow tumor growth and metastasis. Our findings showed marked and unexpected differences between cell autonomous versus systemic deletions of Akt isoforms with respect to tumor initiation and progression. (iv) The inhibition of AMPK is considered pro-tumorigenic. Paradoxically, we found that AMPK activation is required for cell survival during solid tumor formation and possibly metastasis. We will delineate the role of AMPK during metastasis in human cells and mouse models of breast and prostate cancer. We anticipate that the proposed studies will uncover new mechanisms of tumorigenesis associated with cancer metabolism, yield strategies that exploit cancer metabolism for cancer therapy, and uncover vulnerabilities of cancers displaying hyperactivation of PI3K/Akt signaling.